Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / mit-han-lab/temporal-shift-module issues and pull requests

#84 - different num segments in TSM and TSN comparison

Issue - State: closed - Opened by yuvals1 over 4 years ago - 1 comment

#83 - Online model always result in idx to '21'

Issue - State: open - Opened by dancingpipi over 4 years ago - 3 comments

#82 - training with and without TSM correctly

Issue - State: closed - Opened by yuvals1 over 4 years ago - 1 comment

#81 - KeyError: 'state_dict' when trying to run with ResNet50

Issue - State: closed - Opened by Fritskee over 4 years ago

#80 - Pretrained models are not accessible

Issue - State: closed - Opened by nerdysandy17 over 4 years ago - 5 comments

#79 - Could you share online mobilenetv2 pretrained model?

Issue - State: closed - Opened by dancingpipi over 4 years ago - 1 comment

#78 - Error during downloading pre-trained models

Issue - State: closed - Opened by KimManjin over 4 years ago - 3 comments

#77 - Does this pytorch model support TensorRT acceleration?

Issue - State: open - Opened by sunnymoon155 over 4 years ago - 6 comments

#76 - How to train a flow model?

Issue - State: closed - Opened by Zhuysheng over 4 years ago

#75 - Error in online_demo (nvcc missing)

Issue - State: closed - Opened by Gelb153 over 4 years ago - 4 comments

#74 - Low accuracy of Sth-sth v2 pretrained model.

Issue - State: closed - Opened by jd730 over 4 years ago - 18 comments

#73 - Have you ever tried to pretrain TSM-Bniception on Kinetics.

Issue - State: closed - Opened by White1973 over 4 years ago - 1 comment

#72 - About the #Params 24.3M

Issue - State: closed - Opened by Andy1621 over 4 years ago - 1 comment

#71 - Online demo performance tuning

Issue - State: open - Opened by ragavendrams over 4 years ago - 1 comment

#70 - Any pre-training in Kinetics-400 to get 45.6% top-1 in Sth-Sth V1?

Issue - State: closed - Opened by Andy1621 over 4 years ago - 1 comment

#69 - Autotvm warning when running TSM Online demo on Jetson TX2

Issue - State: closed - Opened by ragavendrams over 4 years ago - 1 comment

#68 - Still don't know how to get val_videofolder.txt

Issue - State: open - Opened by zjzhengyin over 4 years ago - 4 comments

#67 - How to train uni-direction TSM ? for example？？

Issue - State: open - Opened by LLLLLLoki over 4 years ago - 2 comments

#66 - I trained a model myself. How do I test it online in real time

Issue - State: open - Opened by LLLLLLoki over 4 years ago - 6 comments

#65 - Pretrained model for sth-sth v1 and kinetics got lower accuracy

Issue - State: open - Opened by Andy1621 almost 5 years ago - 2 comments

#64 - When is inplace shifting going to be supported?

Issue - State: open - Opened by iantimmis almost 5 years ago - 3 comments

#63 - Error when running online demo main.py

Issue - State: closed - Opened by WSPluta almost 5 years ago - 29 comments

#62 - Image resizing

Issue - State: closed - Opened by Polkach almost 5 years ago - 1 comment

#61 - How to finetune the online MobileNetV2 TSM for a Jester like usecase

Issue - State: open - Opened by FabianHertwig almost 5 years ago - 4 comments

#60 - get prediction probability

Issue - State: closed - Opened by sunnymoon155 almost 5 years ago - 1 comment

#59 - error when convert pytorch model to Torch Script

Issue - State: closed - Opened by zhiweizhong almost 5 years ago - 2 comments

#58 - Question about tune-from option

Issue - State: closed - Opened by tilyevsk almost 5 years ago - 1 comment

#57 - Optical flow baselines

Issue - State: open - Opened by willprice almost 5 years ago - 1 comment

#56 - Update README.md

Pull Request - State: closed - Opened by yaoyaoding almost 5 years ago

#55 - Online Demo error

Issue - State: closed - Opened by jaqif almost 5 years ago - 19 comments

#54 - could use the TSM model to get video embedding features ?

Issue - State: open - Opened by ucasiggcas almost 5 years ago - 1 comment

#53 - Training speed about Optical Flow Model.

Issue - State: open - Opened by zhangyuan1994511 almost 5 years ago - 3 comments

#52 - A question about optimizer policy

Issue - State: closed - Opened by ZJUwangdaiyin almost 5 years ago - 1 comment

#51 - Accuracy in the somethingv1 dataset

Issue - State: closed - Opened by anti199994 almost 5 years ago - 4 comments

#50 - Pretrain model on jester dataset ?

Issue - State: closed - Opened by zhangyuan1994511 almost 5 years ago

#49 - code for video object detection

Issue - State: open - Opened by mathfinder almost 5 years ago

#48 - Pretrained model on Kinetics with input of Optical Flow

Issue - State: closed - Opened by misunusc almost 5 years ago - 3 comments

#47 - Training MobilenetV2 from scratch on Kinetics

Issue - State: open - Opened by yifanjiang19 almost 5 years ago - 4 comments

#46 - IndexError: list index out of range???

Issue - State: open - Opened by rgzn-aiyun almost 5 years ago - 3 comments

#45 - where is archs.small_resnet.ResNet?

Issue - State: closed - Opened by ucasiggcas almost 5 years ago - 1 comment

#44 - add target for rasp 4

Pull Request - State: closed - Opened by yaoyaoding almost 5 years ago

#43 - how to process the dataset ?

Issue - State: open - Opened by ucasiggcas almost 5 years ago - 2 comments

#42 - how to test ?

Issue - State: closed - Opened by ucasiggcas almost 5 years ago - 5 comments

#41 - how to test on a movie? mp4 or avi

Issue - State: closed - Opened by ucasiggcas almost 5 years ago - 2 comments

#40 - fix chunking equal number of videos for each thread.

Pull Request - State: open - Opened by pulinagrawal almost 5 years ago - 1 comment

#39 - how to train mobilenetv2 model

Issue - State: open - Opened by 121649982 almost 5 years ago - 9 comments

#38 - the result in the arxiv v1 version is inconsistent with ICCV19 version

Issue - State: closed - Opened by FingerRec almost 5 years ago - 1 comment

#37 - For Uni-directional TSM for online video detection

Issue - State: open - Opened by malarsaravanan1991 almost 5 years ago

#36 - Convert the editing into bash command

Pull Request - State: open - Opened by Michaelvll almost 5 years ago

#35 - 404 to install openCV 4 on Nano

Issue - State: closed - Opened by jiportilla almost 5 years ago - 1 comment

#34 - Question about finetune on UCF101

Issue - State: open - Opened by OliverHxh about 5 years ago - 9 comments

#33 - which dataset uses the pertained model from kinetics?

Issue - State: closed - Opened by beckybai about 5 years ago - 5 comments

#31 - Use my own data

Issue - State: closed - Opened by jwdwzxd about 5 years ago - 4 comments

#30 - run the online_demo by using my trained model

Issue - State: closed - Opened by noahzhy about 5 years ago - 4 comments

#29 - About 3D Network？

Issue - State: closed - Opened by FloydEdwin about 5 years ago - 3 comments

#28 - How long did TSM take to train from scratch on Kinectics ?

Issue - State: closed - Opened by malarsaravanan1991 about 5 years ago - 7 comments

#27 - The model performance of train-from-scratch models

Issue - State: closed - Opened by duducheng about 5 years ago - 1 comment

#26 - 代码哪里可以设置成Offline temporal shift (bi-direction)模式？

Issue - State: closed - Opened by TimeChi about 5 years ago - 1 comment

#25 - Can the model run in mac cpu mode?

Issue - State: closed - Opened by OptimusPrimeCao about 5 years ago - 3 comments

#21 - Script for training TSM on something-something-v2

Issue - State: closed - Opened by pplntech about 5 years ago - 1 comment

#18 - Some doubts about the performance

Issue - State: closed - Opened by deepcs233 about 5 years ago - 2 comments

#17 - How to load model trained by myself to test_models.py

Issue - State: closed - Opened by chenyihang1993 about 5 years ago - 5 comments

#16 - The accuary of TSM-NL is only 57.85%

Issue - State: open - Opened by chenyihang1993 about 5 years ago - 3 comments

#15 - Build Executor... taking time

Issue - State: open - Opened by tomriddle54 about 5 years ago - 2 comments

#14 - Segmentation fault when running demo on ubuntu

Issue - State: closed - Opened by tomriddle54 about 5 years ago - 1 comment

#13 - aarch64: libgomp.so.1: cannot allocate memory in static TLS block

Issue - State: open - Opened by tomriddle54 about 5 years ago - 1 comment

#12 - set params when training the flow model

Issue - State: closed - Opened by arunos728 about 5 years ago - 6 comments

#11 - Training script for online TSM

Issue - State: open - Opened by sleeping1013 about 5 years ago - 11 comments

#10 - [solved problem] for pretrained something-v2 models, it should be "n_round = 2"

Issue - State: closed - Opened by yangwangx about 5 years ago - 2 comments

#9 - The accuracy of TSM finetuned in UCF101

Issue - State: closed - Opened by chenyihang1993 about 5 years ago - 1 comment

#8 - How to train a RGB + Flow model?

Issue - State: closed - Opened by sunnymoon155 about 5 years ago - 3 comments

#7 - Segmentation fault in the online_demo code

Issue - State: closed - Opened by rachillesf about 5 years ago - 9 comments

#6 - questions on online video object detection

Issue - State: closed - Opened by gongbudaizhe about 5 years ago - 3 comments

#5 - Keras implementation of the temporal shift

Issue - State: closed - Opened by thetonus about 5 years ago

#4 - Update main.py

Pull Request - State: closed - Opened by markgreenlqy over 5 years ago

#3 - How to set the params when training the flow model?

Issue - State: closed - Opened by Zhuysheng over 5 years ago - 1 comment

#2 - Add scikit-learn as listed dependency in README

Pull Request - State: closed - Opened by willprice over 5 years ago

#1 - ResNet50 Pretrained Models

Issue - State: closed - Opened by tzzcl over 5 years ago - 1 comment